This proposal describes a 5-year training program to prepare Dr. Ednan Khalid Bajwa for an academic career in molecular epidemiology and ARDS risk and outcomes. Dr. Bajwa received a medical degree from the New York University School of Medicine and will receive an MPH degree from the Harvard School of Public Health (HSPH). He has completed a residency in Internal Medicine at the Massachusetts General Hospital (MGH) in 2003 and will complete his fellowship in Pulmonary and Critical Care Medicine through the Harvard Combined Fellowship Training Program in June 2007. While completing his fellowship Dr. Bajwa has been appointed to the MGH faculty. The program proposed will provide training in laboratory-based techniques and genetic and molecular epidemiology to prepare Dr. Bajwa for a career as an independent clinical researcher. Dr. Bajwa's mentor for this project is Dr. David Christiani, Professor of Medicine at Harvard Medical School and Professor of Occupational Medicine and Epidemiology at HSPH. Dr. Christiani is a world leader in the field of molecular epidemiology. He is the Principal Investigator for multiple NIH- funded R01 grants, and has a well-established record of training investigators for careers in clinical research. This research proposal builds on preliminary data Dr. Bajwa has compiled while working with Dr. Christiani. The goal of the proposed research is to identify genetic and clinical factors that influence the development and pathophysiology of Acute Respiratory Distress Syndrome (ARDS), as well as to evaluate the role of candidate phenotypic markers. Variation in polymorphisms in candidate genes may influence the host response to injurious stimuli, which can be further influenced by clinically modifiable factors such as hyperglycemia. Biomarkers can provide further insight into these processes. This study proposes to utilize a large cohort of patients at risk for ARDS to: (1) determine the role of variation in genes in the PBEF pathway (PBEF, NFKB1, MAPK14, CASP3, MYLK, IL8)in ARDS risk and outcomes; (2) to to determine the role of diabetes, hyperglycemia, and insulin therapy in ARDS risk and outcome, and (3) to determine the role of candidate phenotypic markers in ARDS diagnosis and prognosis. From a public health standpoint, ARDS is a common, highly lethal disease whose pathophysiology is poorly understood and for which there are few therapies that alter prognosis. Identifying patients at highest risk of developing ARDS or having adverse outcomes is critical to improving understanding of this disease and developing effective treatments.